Project Summary Neonatal resuscitation can be challenging for healthcare teams due to its complex nature, requiring a combination of content knowledge, technical skills, behavioral skills, and teamwork. Simulation as a training tool can address this complex skill and also address the infrequent nature of resuscitations in conditions such as very preterm birth or hypoxic ischemic encephalopathy. While simulation is common in neonatal resuscitation training, the effectiveness of using simulation on actual clinical outcomes is unknown. We will test whether using in situ simulation training ? simulation training in the actual setting that healthcare teams practice ? improves team performance and clinical outcomes. This hypothesis will be tested in a stepped wedge trial in a large population-based quality improvement network. A stepped wedge trial allows for all participants to receive the intervention, and therefore increases recruitment ability and gives all participants to benefit from the intervention. In this design, the intervention is rolled out over time with some centers starting earlier than others. Each center is able to then serve as its own control, increasing the power to make comparisons. The setting of this research project will be the California Perinatal Quality Care Collaborative (CPQCC), a population-based network of neonatal intensive care units. CPQCC includes both academic and community units, which means that results will be generalizable. CPQCC already has an existing data infrastructure that includes maternal and neonatal data, including follow-up data at 2 years of age giving an opportunity to study outcomes that does not exist in similar networks. In this project, 40 hospitals will engage in a proven quality improvement model for neonatal resuscitation. In a stepwise fashion, each unit will learn and implement in situ simulation. The setting of the CPQCC allows for a unique opportunity to study the impact of training programs on important clinical outcomes for the most common cause of mortality and long-term morbidity in newborns, preterm birth. In Aim 1, we will assess whether in situ simulation improves clinical outcomes for preterm infants. In Aim 2, we will assess whether better team performance in simulation predicts better clinical outcomes. This will help to inform training methodologies and assessment in neonatal resuscitation. In Aim 3, we will examine facilitators and barriers of implementing simulation training across hospitals. This will inform the implementation of training programs across hospitals to improve neonatal resuscitation. In Aim 4, we will evaluate perform an economic evaluation of in situ simulation and implementation. The results of this research will provide a clearer understanding of how simulation can be used as an assessment tool, and more importantly the impact that it can have on improving clinical outcomes.